SupportingInformation(SI)forJournalofMaterialsChemistryA

Carbohydrate based hyper-cross-linked organic polymers with –OH functional group for CO2 separation†

Haiying Li,a, b# Bo Meng,c# Shannon M. Mahurin,b Songhai Chai,c Kimberly M. Nelson,c David C. Baker,c Honglai Liua* and Sheng Daib,c*

a State Key Laboratory of Chemical Engineering and Department of Chemistry, East China University of Science and

Technology, Shanghai, 200237, China.

Email: hlliu@ecust.edu.cn b Chemical Science Division, Oak Ridge National Laboratory, Oak Ridge, TN, 37831, United States.

Email: dais@ornl.gov c Department of Chemistry, University of Tennessee, Knoxville, TN 37996, United States # These authors contributed equally to this work.

1. Experimental details 1.1 General carbohydrate monomer synthesis All chemicals were purchased as reagent grade and used without further purification, unless

otherwise noted. Reagent grade dichloromethane (DCM), tetrahydrofuran (THF), methanol

(MeOH) and N, N-dimethylformamide (DMF) were obtained from the Pure-Solv (Innovation

Technologies) solvent system that uses alumina columns, except for DMF, which was dried over

a column of 5Å molecular sieves. Pyridine was distilled over CaH2 prior to use. All reactions

were performed under anhydrous conditions unless otherwise noted. Reactions were monitored

by thin-layer chromatography (TLC) on silica gel precoated aluminum plates. Zones were

detected by UV irradiation using a 254 nm lamp and/or by heat/charring with p-anisaldehyde–

sulfuric acid development reagent. Column chromatography was performed on silica gel (40−63

μm). 1H and 13C NMR spectra were recorded at room temperature with a Varian VNMRS 500

MHz or a Varian VNMRS 600 instrument. Chemical shifts are reported in δ-units (ppm) relative

to the residual 1H CDCl3 at δ 7.26 ppm and 13C at δ 77.16 ppm. Mass spectrometric analysis was

performed on a QSTAR Elite quadrupole time-of-flight (QTOF) mass spectrometer with an ESI

source. Compounds Glc-3 and Ara-1 were purchased from Sigma−Aldrich and compound Gal-1

was purchased from Santa Cruz Biotechnology. Compounds Glc-1,S1 Glc-2,S2 and Gal-2S2 are

synthesized on the base of literature procedures.

Electronic Supplementary Material (ESI) for Journal of Materials Chemistry A.This journal is © The Royal Society of Chemistry 2015

Methyl 2,3,4,6-tetra-O-benzyl-α-D-glucopyranoside (Glc-1): Methyl α-D-glucopyranoside (500

mg, 2.57 mmol) was dissolved in dry DMF (10 mL). The solution was cooled to 0 °C and added

NaH (60%, 824 mg, 20.60 mmol) portionwise. After stirring for 15 min, benzyl bromide (1.84

mL, 15.45 mmol) was slowly added and the reaction mixture was warmed to room temperature.

After stirring overnight, the reaction mixture was quenched by adding methanol (5 mL) and

concentrated in vacuo. The resulting residue was mixed with water (100 mL) and extracted with

methylene chloride (3 × 50 mL). The combined organic layers were dried over anhydrous Na2SO4

and concentrated. The crude product was purified by silica gel chromatography

(Hexanes:EtOAc=15:1 to 7:1) to yield Glc-1 (1.24 g, 87%) as colorless oil. 1H NMR (CDCl3, 600

MHz) δ 7.42 – 7.20 (m, 20H), 5.06 (d, J = 10.9 Hz, 1H), 4.90 (d, J = 5.1 Hz, 1H), 4.89 (d, J = 5.2

Hz, 1H), 4.86 (d, J = 12.1 Hz, 1H), 4.73 (d, J = 12.1 Hz, 1H), 4.70 (d, J = 3.5 Hz, 1H, H-1), 4.67

(d, J = 12.2 Hz, 1H), 4.55 (d, J = 3.1 Hz, 1H), 4.53 (d, J = 4.4 Hz, 1H), 4.06 (t, J = 9.3 Hz, 1H),

3.82 (m, 1H), 3.78 (dd, J = 10.5, 3.8 Hz, 1H), 3.72 – 3.68 (ddd, J = 10.0, 5.3, 3.2 Hz, 2H), 3.63

(dd, J = 9.7, 3.5 Hz, 1H), 3.44 (s, 3H). 13C NMR (CDCl3, 151 MHz) δ 138.92, 138.38, 138.29,

138.04, 128.58 – 127.71 (Ar-C), 98.37, 82.29, 79.99, 77.82, 75.89, 75.21, 75.13, 73.61, 70.21,

68.60, 55.32. HRMS ESI (m/z) (M+Na)+ calcd for C35H38O6Na+ 577.2566, found 577.2566. The

data of Glc-1 are consistent with those previously reported.S1

Methyl 2,3,4-tri-O-benzyl-α-D-glucopyranoside (Glc-2). A solution of Methyl α-D-

glucopyranoside (400 mg, 2.06 mmol) in anhydrous pyridine (10 mL) was added imidazole (280

mg, 4.12 mmol) and cooled to 0 °C. Then tert-butylchlorodimethylsilane (372 mg, 2.47 mmol)

was added, and the mixture was stirred at room temperature for 24 h. The reaction mixture was

quenched by adding water and concentrated in vacuo. The crude product was purified by silica

gel chromatography (Hexanes:EtOAc:CH2Cl2:MeOH=6:1:1:0.5) to afford methyl 6-O-(tert-

butyldimethylsilyl)-α-D-glucopyranoside (527 mg, 83%) as a white solid. It was then dissolved in

anhydrous DMF (10 mL) and cooled to 0 °C. The solution was treated with NaH (60%, 547 mg,

13.68 mmol) portionwise and stirred at 0 °C for 15 min. Benzyl bromide (1.2 mL, 10.26 mmol)

was slowly added and the reaction mixture was warmed to room temperature. After stirring

overnight, the reaction mixture was quenched by adding methanol (5 mL) and concentrated in

vacuo. The resulting residue was mixed with water (100 mL) and extracted with methylene

chloride (3 × 50 mL). The combined organic layers were dried over anhydrous Na2SO4 and

concentrated. The crude product was purified by silica gel chromatography

(Hexanes:EtOAc=20:1 to 8:1) to yield methyl 2,3,4-tri-O-benzyl-6-O-(tert-butyldimethylsilyl)-α-

D-glucopyranoside (690 mg, 70%) as a white solid. It was then dissolved in anhydrous THF (10

mL) and added tetrabutylammonium fluoride solution (1.0 M in THF, 2.39 mmol). The resulting

mixture was stirred at room temperature for 16 h. After concentration, the mixture was quenched

by water (100 mL) and extracted with methylene chloride (3 × 50 mL). The combined organic

layers were dried over anhydrous Na2SO4 and concentrated again. The crude product was purified

by silica gel chromatography (Hexanes:EtOAc=3.5:1) to yield Glc-2 (431 mg, 78%) as colorless

oil. 1H NMR (CDCl3, 600 MHz) δ 7.41–7.31 (15 H, m), 5.04 (d, J = 10.9 Hz, 1H), 4.93 (d, J =

11.0 Hz, 1H), 4.89 (d, J = 11.0 Hz, 1H), 4.84 (d, J = 12.1 Hz, 1H), 4.71–4.69 (dd, J = 11.6, 8.5

Hz, 2H), 4.62 (d, J = 3.5 Hz, 1H, H-1), 4.06 (m, 1H), 3.81 (dd, J = 11.8, 2.6 Hz, 1H), 3.74 (dd, J

= 11.8, 4.1 Hz, 1H), 3.69 (ddd, J = 10.0, 4.1, 2.7 Hz, 1H), 3.59 (m, 1H), 3.53 (dd, J = 9.7, 3.6 Hz,

1H), 3.40 (s, 3H), 1.96 (s, 1H). 13C NMR (CDCl3, 151 MHz) δ 138.77, 138.18, 138.14, 128.50,

128.43, 128.14, 128.04, 127.99, 127.97, 127.88, 127.64, 98.19, 81.98, 80.01, 77.43, 75.77, 75.05,

73.43, 70.75, 61.80, 55.21. HRMS ESI (m/z) (M+Na)+ calcd for C28H32O6Na+ 487.2097, found

487.2094. The data of Glc-2 are consistent with those previously reported.S2

Methyl 2,3,4-tri-O-benzyl-α-D-galactopyranoside (Gal-2). A similar synthetic procedure for

preparing Glc-2 was used. 1H NMR (500 MHz, Chloroform-d) δ 7.43 – 7.29 (m, 15H) 5.00 (d, J

= 11.6 Hz, 1H), 5.00 – 4.90 (m, 2H), 4.78 (d, J = 11.7 Hz, 1H), 4.72 (m, 2H), 4.67 (d, J = 11.6

Hz, 1H), 4.08 (dd, J = 10.1, 3.6 Hz, 1H), 3.96 (dd, J = 10.1, 2.8 Hz, 1H), 3.89 (dd, J = 2.9, 1.0 Hz,

1H), 3.73 (m, 2H), 3.50 (m, 1H), 3.38 (s, 3H), 1.79 (s, 1H). 13C NMR (126 MHz, cdcl3) δ 138.82,

138.53, 138.28, 128.67, 128.57, 128.54, 128.47, 128.19, 128.08, 127.85, 127.72, 127.66, 98.93,

79.22, 76.58, 75.20, 74.55, 73.72, 73.69, 70.37, 62.49, 55.46. HRMS ESI (m/z) (M+Na)+ calcd

for C28H32O6Na+ 487.2097, found 487.2095. The data of Gal-2 are consistent with those

previously reported.S2

1H NMR Methyl 2,3,4,6-tetra-O-benzyl-α-D-glucopyranoside (Glc-1)

13C NMR Methyl 2,3,4,6-tetra-O-benzyl-α-D-glucopyranoside (Glc-1)

1H NMR Methyl 2,3,4-tri-O-benzyl-α-D-glucopyranoside (Glc-2)

13C NMR Methyl 2,3,4-tri-O-benzyl-α-D-glucopyranoside (Glc-2)

1H NMR Methyl 2,3,4-tri-O-benzyl-α-D-galactopyranoside (Gal-2)

13C NMR Methyl 2,3,4-tri-O-benzyl-α-D-galactopyranoside (Gal-2)

1.2 General polymerization of benzylated carbohydrates Typically, to a solution of the monomer and FDA in anhydrous 1,2-dichlorehane (10mL), a slurry of FeCl3 in DCE (10mL) was slowly added under nitrogen atmosphere. The mixture was then heated to 45C for 5h and 80C for 19h. The resulting brown precipitate was collected and washed with methanol and water until the filtrate became colorless and further purified by Soxhlet Extraction with methanol for 24h. The polymer was dried under vacuum for 24h at 60C. For different monomers, the ratio of external cross-linker FDA and catalysis FeCl3 was adjusted because of the diverse numbers of benzyl rings as illustrated in Table S1.

Table S1 composition and porosity of samples

FDAa FeCl3a Smicro(m

2/g)b Vtotal (cm3 g-1)c

Vmicro (cm3 g-1)d

Smicro/SBE

T Vmicro/Vtot

al

Glc-1 8 8 445 0.40 0.20 60.5% 50% Glc-2 6 6 456 0.47 0.21 55.9% 44% Glc-3 8 8 479 0.47 0.22 57.7% 46% Gal-1 8 8 482 0.47 0.22 56.1% 46% Gal-2 6 6 314 0.93 0.14 28.8% 15% Ara-1 2 2 237 0.30 0.11 50.4% 35%

a Molar ratio with respect to monomer. b Micropore surface area calculated form the nitrogen isotherm at P/Po= using T method. c Pore volume calculated from nitrogen isotherm at P/Po=0.889. d Micropore volume calculated form the nitrogen isotherm using T-method.

Table S2 the BET surface area of all the samples after water treatment (PH=5, 80°C, 48h)

SBET(m2/g) SBET after treatment

(m2/g)

SBET percentage (%)

Vmicro (cm3 g-1)

Vmicro after treatment (cm3 g-1)

Vmicro percentage (%)

Glc-1 735 672 91.4 0.20 0.18 92 Glc-2 816 702 86.1 0.21 0.18 87 Glc-3 829 753 90.9 0.22 0.21 94 Gal-1 858 801 93.4 0.22 0.21 97 Gal-2 1090 1054 96.7 0.14 0.14 99 Ara-1 470 410 87.3 0.11 0.09 84

1.3 Characterization of polymers

Figure S1. 13C solid state NMR with spinning rate of 7k and FTIR spectrums for all carbohydrate HCPs.

Figure S2. X-ray powder diffraction (XRD) spectra of all carbohydrate polymers.

Figure S3. Thermal gravimetric analysis (TGA) analysis of all carbohydrate polymers heated at a rate of 10C/min up to 900C under nitrogen flow. Isothermal adsorption analysis was performed using single gas adsorption isotherms of CO2 and N2. The adsorption isotherms were fit using the Toth equation:S3

For complete monolayer adsorption coverage, nio is introduced in the modified Langmuir

as the amount adsorbed (mmol g-1), qsat for the maximum capacity, b is the temperature dependent affinity parameter, t is fitted as a measure of heterogeneity of the surface, all as a function of pressure (P, kPa). By fitting the isotherms, we can obtain the lnP vs. 1/T at constant loading for each temperature. The slope corresponds to the heat of adsorption, -ΔH (kJ mol-1), as a function of loading (mmol g-1). Isotherm fits of absolute loading are used for ideal adsorbed solution theory (IAST) developed by Myers and Prausnitz.S4 MatLab® was used to calculate the adsorption selectivity from the single gas isotherms of N2 and CO2 at a molar ratio of 0.85 : 0.15, respectively.

nio qsatbP

[1 (bP)t ]1/t

FigurN2 at

re S4. Adsorpt 273 K and 29

tion (solid sym98 K respectiv

mbols) and devely.

esorption (openn symbols) isootherms of all samples for C

CO2 and

Figure S5. Initial slope calculations for CO2 and N2 isotherms collected at 273K and 298 K respectively.

Figure S6. Mixed gas selectivities calculated using the ideal adsorbed solution theory for all samples under 0.15mol CO2 and 0.85mol N2 versus pressure. Table S3. Mixed gas selectivities calculated for all samples under 0.15mol CO2, 0.85mol N2 at select pressures. Glc-1 Glc-2 Glc-3 Gal-1 Gal-2 Ara-1 Selectivities at 10kPa

4 24 35 12 32 6

Selectivities at 20kPa

6 28 44 15 35 8

Selectivities at 100kPa

26 44 96 31 48 23

Table S4 the comparison of surface area, CO2 uptake, selectivity (CO2/N2) (at 273 and 298 K) and isosteric heat (Qst) in selected POPs with –OH functional groups.

MOP SBET (m2g-

1) CO2 uptake (mmol g-1)

T (K) Selectivity Qst (kJ mol-1)

Ref.

Glc-3 829 2.43 273 41 25.8 This work 1.45 298 27 This work

1-naphthol 414 1.85 273 28-31 S5 1.25 298 16 S5

1,1-bi-2-naphthol 1015 3.96 273 28-31 S5 2.27 298 26 S5

phenol 400 2.14 273 - - S6 Tetraphenylethylene-

HCP 618 1.92 273 119 - S7

1.12 298 - S7

2. Computational details Quantum chemistry calculations were performed based on DFT method by G09 program. Since

the calculation processes are expensive and time-consuming, basic monomers with simplified structures were applied in the assessment of each type of hyper cross-linked polymers (Figure S6). The geometries of all the monomer and monomer−gas pairs were fully optimized with M062x/6-311g(d, p) method, which is one of the most popular DFT methods in the study of intermolecular interactions. Frequency calculations at the same level were also performed to make sure all optimized structures can represent the actual minimum potential energy. Moreover, all the interaction calculations were corrected with the basis set superposition error (BSSE) by Boys and Bernardi’s procedure. In the initial analysis of CO2−carbohydrate pair systems’ geometries, both hydrogen bonding and dipole-quadrupole interactions were considered as initial structures(Figure S7). The binding energy between carbohydrate and gases was calculated by the following formula:

ΔE = E(gas−monomer)-E(gas)-E(monomer)+E(BSSE).

Figure S7. Simplified structures of calculated carbohydrate monomers.

Figure S8. Initial structures of simplified carbohydrate monomer-gas dimer for optimization. All the structures of monomers and gas molecules used for dimer optimization were fully optimized at the same level first. References:

S1. G.J.L. Bernardes, E.J. Grayson, S. Thompson, J.M. Chalker, J.C. Errey, F.E. Oualid, T.D.W.

Claridge, B.G. Davis, Angew. Chem. Int. Ed., 2008, 47, 2244-2247.

S2. E. Sasaki, C-I Lin, K-Y Lin, H-W Liu, J. Am. Chem. Soc., 2012, 134, 17432-17435.

S3. D.D. Do. (1998). Adsorption Analysis: Equilibria and Kinetics. London, Imperial College

Press.

S4. A.L. Myers, J.M. Prausnitz, "Thermodynamics of Mixed-Gas Adsorption " AIChE J., 1965,

11(1), 1-7.

S5. R. Dawson, L. Stevens, T. C. Drage, C. E. Snape, M. W. Smith, D. J. Adams, A. I. Cooper, J. Am. Chem. Soc. 2012, 134, 10741.

S6. B. Y. Li, R. N. Gong, W. Wang, X. Huang, W. Zhang, H. M. Li, C. X. Hu, B. E. Tan, Macromolecules 2011, 44, 2410.

S7. S. Yao, X. Yang, M. Yu, Y. Zhang and J. Jiang. J. Mater. Chem. A, 2014, 2, 8054–8059.